Weekend-free and robotics-compatible protocols for high-throughput human-induced pluripotent stem cell maintenance, amplification, and differentiation

Abstract

Background: Human induced pluripotent stem cells (hiPSCs) are essential tools for disease modeling, drug testing, and regenerative medicine. These applications require differentiation of multiple cell lines in parallel, synchronized production of large numbers of hiPSCs and differentiated cells, and assessment of differentiation efficiency before proceeding to in-depth characterization through molecular and cellular assays. However, hiPSC culture protocols are often laborious manual processes, which affects reproducibility and makes high-throughput applications challenging.

Methods: We describe high-throughput hiPSC maintenance and amplification protocols that are weekend-free, scalable, and can be performed manually or automated. These optimized protocols enable scale-up of hiPSC production and differentiation to multiple cell types, including endothelial cells, microglia, and retinal pigment epithelial cells.

Results: We characterized the differentiated cells using molecular and cellular assays and demonstrated that cells generated with this approach represent accurate cell identities.

Conclusion: The approaches described here enable high-throughput hiPSC applications and improve reproducibility and scalability.If you want, I can also tighten this further to better fit a strict 350-word journal limit or adapt the heading style to a specific journal.

Graphical Abstract

Graphical Abstract.

Significance statement.

In this study, we established protocols for hiPSC amplification and maintenance that are weekend-free without sacrificing their differentiation capacity. The protocols are suitable for high-throughput maintenance, amplification, and differentiation, with and without automation. These protocols were applied to expand a cohort of 72 hiPSC lines in less than 6 months. The weekend-free, scalable protocols described here overcome the bottlenecks of labor-intensive manual processes and reduce batch variability, thus improving consistency and scale of hiPSCs.

Introduction

The discovery of human induced pluripotent stem cells (hiPSCs) by Yamanaka and colleagues in 2007 was a groundbreaking achievement that has transformed the field of stem cell research.¹ Since then, hiPSCs have emerged as valuable cellular models for disease modeling, drug testing, toxicity screening, and regenerative medicine.^2,3 HiPSCs provide the unique ability to differentiate into various cell types, allowing the generation of multiple disease-relevant cell lines that can be instrumental in studying the underlying mechanisms of various disorders, including cardiovascular, neurodegenerative, and metabolic diseases.^4–18

Applying functional genomics, transcriptomic analysis, and a wide range of functional assays in hiPSC-derived models have significantly enhanced our understanding of both monogenic and complex diseases,¹⁹ the latter is driven by a combination of environmental factors and multiple genetic variants.^2,20 Nevertheless, many of these studies rely on a relatively small number of hiPSC patient lines, limiting its reproducibility and the ability to fully understand underlying disease biology.²¹ This highlights the importance of increasing the sample size and the diversity of hiPSC lines to study complex diseases.²¹

The potential of hiPSCs in advancing medical research is immense, but utilizing this potential comes with its own challenges. Manual culturing of hiPSCs and their subsequent differentiation into specific cell types is a complex and extremely laborious process that requires extensive knowledge and training,²² daily medium changes, and quality control assessments.^23,24 As the cell population expands, the workload increases exponentially, making it challenging to scale up the production of hiPSCs.²⁵ Furthermore, manual hiPSC culture can potentially introduce a higher risk of operator bias, such as contamination or inconsistent culture conditions, leading to variations in cell growth and differentiation potential, not only impacting the quality and reliability of the hiPSCs but further hindering the scalability. Batch-to-batch variability can also be introduced during the differentiation process, with hiPSC lines responding differently to the same differentiation protocol, leading to variations in the results. Additionally, culture conditions can also impact the differentiation potential of hiPSCs, such as the use of feeder cells or the type of growth factors employed in the differentiation process.^18,26–29

Continuous efforts are being made in the hiPSC field to address these challenges. In recent years, research groups have started to implement the use of robotic systems to increase the throughput and efficiency of hiPSC reprogramming, maintenance, and differentiation.^25,30 Several studies have demonstrated the efficient use of CompacT SelecT, an integrated robotic system that can automate the entire cell culture process. CompacT SelecT systems have been efficiently used for different types of differentiations, including hepatocytes, cardiomyocytes, and neurons.^25,30–33 However, machines such as robots can encounter malfunctions at any time, including over weekends. Hence, further improvements to minimize any kind of operational disruptions are needed to overcome the challenges associated with lengthy protocols that require work over the weekend.

Scientists are continuously exploring the development of more streamlined protocols to reduce the time and the effort required for hiPSC maintenance while simultaneously maintaining high quality and scalability.³⁴ A study by Kuo et al.³⁵ demonstrated that by using chemically defined E8 media and specific cell splitting ratios, it was possible to maintain hiPSCs following a weekend-free protocol. However, challenges remain in consistently applying these ratios across different hiPSC lines with varying growth rates. Variability in growth rate can lead to dropout of slower-growing cell lines in experiments where multiple hiPSCs are grown together in one dish to model population-scale hiPSC studies.^{19,25,30–33,36}

This study aims to address these gaps by developing an efficient weekend-free, high-throughput hiPSC maintenance protocol optimized for use with the automated CompacT SelecT system, and which would be feasible to use for any potential population-scale studies where the growth rate of cell lines is an important factor to consider. We optimized this weekend-free protocol by testing multiple seeding densities in cell culture plates as well as flasks and benchmarked them against conventional manual passage protocols. We show that our weekend-free protocol is highly compatible with manual hiPSC cell culture protocols and can be adapted to an automated robotic handling system. We further used our validated weekend-free automation-compatible protocols to amplify a cohort of 72 hiPSC lines for use in disease modeling and drug testing experiments. After the successful establishment of this shortened maintenance protocol, we further used our protocols to successfully differentiate hiPSC lines into endothelial, microglia, and retinal pigment epithelial cells (RPE). Our results demonstrate the effectiveness of our weekend-free protocol for hiPSC maintenance and differentiation, whether performed manually or using the automated CompacT SelecT system. This showcases the development of a scalable, high-throughput hiPSC cell culture protocol that significantly improves both efficacy and reproducibility in hiPSC production for both research and therapeutic applications.

Results

Establishment and implementation of a weekend-free, robotics-compatible hiPSC passaging protocol

HiPSC culture maintenance is typically a lengthy process that requires proper attention with daily media changes, including on the weekends. In this study, we established a weekend-free hiPSCs passaging schedule (Figure 1A and B) that requires seeding of hiPSCs on Monday and Thursday, followed by media changes on Tuesday and Friday, respectively. Multiple cell seeding and media volume optimizations were performed to ensure that hiPSCs did not overgrow and remained viable, specifically over the weekend. Furthermore, to obtain reproducible results, we minimized cell clumping and obtained single cells by replacing EDTA with Accutase as the cell dissociation reagent. The details of seeding densities, volume of media, and passaging ratios are outlined in Figure 1C for culturing hiPSCs in different types of culture plates. As hiPSCs are not very tolerable to single-cell passage, we also added ROCK inhibitor Y-27632 (ROCKi) at 10 µM to the culture media and removed it the next day. ROCKi has been added to culture media even in colony passaging using EDTA, with no remarkable effect on the cells.³⁵ To adapt the weekend-free protocol to the robotic system, we made some minor modifications to the manual protocol. The schematic protocol used for manual and robotic hiPSC passaging is outlined in Figure 1B.

Figure 1.

Weekend-free passaging schedule that is compatible with robotics. (A) Weekly passaging schedule for split and feed of cell cultures from Monday to Sunday. (B) Detailed splitting protocol for cells, including the washing, incubation, dilution, centrifugation, counting, seeding, and media change steps for daily media change protocol, weekend-free manual, and weekend-free automated, respectively. (C) Table outlining the seeding, feeding, and splitting guidelines for different culture vessels (6-well plate, T25 flask, T75 flask, T175 flask, and triple flask). The table includes the details on culture area, Accutase volume, cell count for seeding, media volume with and without ROCKi, and Matrigel volume per cm². Refer to panel (C) for densities.

Benchmarking of our weekend-free manual and robotic protocol to a conventional daily media change protocol

To validate our weekend-free protocol, we benchmarked the hiPSC cell characteristics to a conventional, daily medium change protocol. Two different hiPSC lines, hiPSC-line 1 and hiPSC-line 2, cultured for at least 50 passages using our weekend-free protocol exhibited the same morphological features in comparison to the daily media change protocol (Figure 2A). In addition, both hiPSC lines maintained their normal karyotype (Figure 2B) and pluripotency as measured by the expression of undifferentiated pluripotency-associated markers NANOG and POUF51 using RT-qPCR (Figure 2C), fibroblasts (ATCC) were added as a control. We also observed consistent immunostaining of pluripotent markers OCT4, TRA-1-60, SOX2, and SSEA4 in both the manual and automated weekend-free protocol in comparison to the daily media change protocol (Figure 2D) across hiPSC-line 1 and hiPSC-line 2.

Figure 2.

Characterization of weekend-free grown hiPSCs for both manual and automated protocols. (A) Phase-contrast images (original magnification 10×) of hiPSC lines. (B) A whole genome view displays all somatic and sex chromosomes in one frame with a high-level copy number for hiPSC line-1 and hiPSC line-2. KaryoStat+ assay was used for the detection of chromosomal abnormalities in hiPSC lines. The pink, green, and yellow colors indicate the raw signal for each individual chromosome probe, while the blue signal represents the normalized probe signal, which is used to identify copy numbers and aberrations (if any). (C) Relative mRNA expression levels for POUF51 and NANOG. Values are normalized first to 18S and then to fibroblasts, with fibroblasts serving as the baseline. Data are presented as mean ± SEM. (n = 3, collected from 3 independent experiments). (D) Immunostaining of several pluripotency markers in hiPSC-line-1 and hiPSC-line-2. Scale bar: 100 µm.

Next, we evaluated the multilineage differentiation potential of the two hiPSC lines using weekend-free manual and automation protocols in comparison to the manual daily media change protocol. A commercial trilineage kit was used to differentiate both hiPSC lines into 3 germ layers: ectoderm, endoderm, and mesoderm (Figure 3A). Gene expression levels for ectoderm markers (NESTIN and PAX6), endoderm markers (FOXA2 and SOX17), and mesoderm markers (CXCR4 and HAND1) were measured across both hiPSC-line 1 and hiPSC-line 2 using RT-qPCR. Both hiPSC lines grown using our weekend-free manual and robotic protocols showed similar differentiation capabilities for all the germ layer markers compared to the daily media change protocol (Figure 3B). Generally, both lines showed similar expression levels for all genes among 3 conditions, with a trend of higher expression levels in weekend-free and automation conditions. In addition to RT-qPCR, we also examined the protein expression for ectoderm markers (PAX6 and SOX9), endoderm markers (FOXA2 and SOX17), and mesoderm markers (TBXT and HNFα) across the two hiPSC lines and observed similar expression patterns across the 3 different protocols (Figure 3C). These benchmarking experiments suggest that our weekend-free protocols are as effective as the conventional daily media change protocol; hence, for the remaining experiments, we decided to culture hiPSCs using our manual and automated weekend-free protocols.

Figure 3.

Characterization of multilineage differentiation of two hiPSC lines grown using weekend-free manual and automated protocols. (A) Trilineage differentiation protocols as described by the manufacturer (7 days of differentiation for ectoderm and 5 days of differentiation for endoderm and mesoderm). (B) Relative mRNA expression levels for ectoderm (NESTIN and PAX6), endoderm (FOXA2 and SOX17), and mesoderm (CXCR4 and HAND1). Values are normalized first to 18S and then to hiPSC at D0, with hiPSC at D0 serving as the baseline. Data are presented as mean ± SEM. (n = 3, collected from 3 independent experiments). (C) HiPSC line-1 and hiPSC line-2 differentiated into ectoderm, endoderm, and mesoderm separately and immunostained for ectoderm markers (PAX6 and SOX9), and endoderm markers (FOXA2 and SOX17), and mesoderm markers (TBXT and HNFα). Scale bar: 100 µm and 50 µm as indicated in the figures.

Weekend-free manual and robotic protocol increase efficiency and scalability in amplification of hiPSC lines for biobanking

Accessibility to a substantial number of patient-derived hiPSC lines is critical in population-based studies to increase our understanding of disease biology and to better capture genetic diversity within populations. For one study, we acquired patient-derived hiPSC lines from a large cohort of 72 patients with age-related macular degeneration from New York Stem Cell Foundation (NYSCF). To maximize the utility of these 72 patient-derived hiPSC lines, it was imperative to maintain an ample supply of these lines by amplifying them through cell passaging and biobanking them for subsequent research applications. We revived the patient-derived hiPSC lines in a single well of a 6-well plate from vials containing 400 000 to 300 000 cells each with viability ranging between 80% and 92%, and then continuously cultured them for 4 passages by transferring them to a new 6-well plate and thereafter to T175 flasks, thereby increasing the overall cell number. We first applied a manual daily medium-change protocol to the initial 20 of 72 hiPSC lines (Figure S1A, see online supplementary material for a color version of this figure). For the subsequent 20 lines, and eventually for the remainder of the cohort, we implemented several workflow changes: we replaced CloneR with ROCKi for consistency, adopted a weekend-free culture schedule, fully leveraged the thermo-stable FGF2 in mTeSR Plus, and introduced automation early in the process (Figure S1B, see online supplementary material for a color version of this figure). Following these changes, we routinely recovered approximately fourfold more cells and were able to freeze more vials per line. This increase in efficiency and scale likely reflects the combined benefits of automation, including a more consistent schedule and handling, although improved performance of ROCKi and the accumulation of secreted factors that may have enhanced growth and survival could also have contributed. In addition, we were able to process 20 lines simultaneously on the robot, spanning 18 days to complete the whole amplification process. To further validate the accuracy of our method, we karyotyped multiple lines from this repository; we are showing data from hiPSC-line 3 and hiPSC-line 4, which both showed normal karyotypes (Figure S1C and D, see online supplementary material for a color version of this figure).

Scalable and automated hiPSC differentiation to multiple cell types ranging from simple and short protocols to long and labor-intensive protocols

After optimizing our maintenance and amplification procedures, our next challenge was to standardize hiPSC maintenance for various differentiation protocols. Our strategy was as follows: (1) We first implemented each differentiation protocol in-house, adhering to the original formats and maintenance conditions described in the literature. (2) In the next phase, we transitioned to our preferred maintenance media for automation, specifically mTeSR Plus with Matrigel, for greater convenience and compatibility with robotic handling. (3) Finally, we adapted the culture format to T175 flasks to support scalability. At each stage, we ran the differentiation protocols through to completion, assessed the efficiency of differentiation, and compared these results to the original protocols (data not shown). This stepwise approach ensured protocol robustness.

For the following differentiations, hiPSCs were either manually maintained and differentiated or both maintained and differentiated using the CompacT SelecT. In the automation protocols, we expanded and differentiated hiPSCs using the CompacT SelecT. We then compared manual and automated protocols for each type of differentiation we did.

Scalable hiPSC differentiation to endothelial cells

To demonstrate the efficiency and the reproducibility of our weekend-free scalable maintenance protocols (manual and robotics), we differentiated two hiPSC lines into endothelial cells using a previously published protocol.³⁷ We started the endothelial differentiations on Fridays after passaging them on Thursdays, providing uniform maintenance and differentiation. The differentiation protocol starts with a mesodermal induction, followed by endothelial specification and maturation (Figure 4A). Endothelial cells derived from two different hiPSC lines using both manual and automated procedures exhibited standard endothelial-like morphological characteristics (Figure 4B) and similar gene expression patterns for multiple endothelial markers like CDH5, PECAM1, and NOTCH4 as measured by RT-qPCR (Figure 4C). Both hiPSC lines (1 and 2) cultured using manual and automated differentiation protocols showed similar expression profiles for the endothelial-specific markers CDH5, PECAM1, and NOTCH4, albeit there were minor differences in expression levels of CDH5 and NOTCH4 between the two hiPSC lines, which may be cell-line intrinsic. We further characterized hiPSC-line-1 using flow cytometry to evaluate protein expression levels of endothelial-specific markers CD31/CD144 by flow cytometry (Figure 4D). Endothelial cells differentiated from hiPSC-line-1 both, manually and automated, as well as HUVEC cells, showed similar expression of CD31 and CD144. Orange histograms showed similar cellular staining for both CD31 and CD144 conjugated to IgG2a in all conditions. IgG2a was used as an isotype control antibody (grey histogram) to assess nonspecific binding. Similarly, endothelial cells differentiated from hiPSC-line-1 and hiPSC-line-2 in manual and automated protocols, along with HUVEC, were further characterized for protein expression of several endothelial-specific markers, including: CD31 with vWF/CLAUDIN5/CD144/ZO1in presence of DAPI nuclear staining using the Operetta system (Figure 4E). Immunostaining data demonstrated highly analogous expression of all endothelial-specific protein markers in both hiPSC lines differentiated using manual and robotic systems compared to the control HUVEC cells. Lastly, we evaluated the functionality of the differentiated endothelial cells using an angiogenesis assay that measures the ability of these cells to form capillary-like tubular structures on an extracellular matrix support, a distinctive property of endothelial cells. The endothelial cells were cultured on Matrigel-coated vessels, we observed similar tubular-forming characteristics of endothelial cells differentiated manually and using robotics (Figure 4F). Interestingly, further quantification of the number of nodes formed, the area of the meshes formed, and the length of the branches showed a significant improvement in the network parameters of vascular-like structures in the automated version of the endothelial differentiation for both hiPSC lines compared to the manual version (Figure 4G). Moreover, this differentiation protocol provides at least 3 freezing points (one at day 5 and at each passage until endothelial cells go into senescence) and, coupled with our weekend-free protocols, provide an innovative and flexible approach for endothelial cell differentiation.

Figure 4.

Characterization of endothelial-like cells differentiated using both manual and automated protocols. (A) Overview of endothelial differentiation protocol used. (B) Phase-contrast image showing typical endothelial-like cells compared to HUVEC (Passage 3, day 4). Scale bar, 400 μm. (C) Relative mRNA expression levels of CDH5, PECAM1, and NOTCH4 in hiPSC at D0 and differentiated hiPSC-derived cells. Values are normalized first to 18S and then to hiPSC at D0, with hiPSC at D0 serving as the baseline. Data are presented as mean ± SEM with individual values indicated by different symbols. Groups include hiPSC-line 1 (manual and automated protocols) and hiPSC-line 2 (manual and automated protocols) (n = 3, collected from 3 independent experiments). (D) Flow cytometry analyses of endothelial markers CD31 and CD144 (n = 3, collected from 3 independent experiments). (E) Endothelial-like cells on day 26, immunostained for CD31, CD144, ZO1, vWF, and CLAUDN5. Scale bar, 50 µm. (F) Endothelial-like cells forming vascular tube–like structures on thick Matrigel layers. Images were obtained 24 h after seeding. Scale bar = 646 µm at 10× magnifications, S3 incucyte. (G) Angiogenesis parameters analyzed with ImageJ: number of nodes formed, length of branches formed, and area of the meshes. Values are the mean and SEM calculated from triplicates in 3 independent experiments (n = 3, *P < .05; **P < .01).

Scalable hiPSC differentiation to microglia-like cells

Next, we aimed to explore the potential of robotics for more complex differentiation protocols. HiPSC differentiation to microglia, the resident immune cells of the central nervous system, consists of multiple complex steps, all of which we successfully automated using the CompacT SelecT. We passaged our cells to start the protocol at D0. We then generated macrophage precursors, which were further differentiated into microglia-like cells (Figure 5A). The microglia differentiated from two hiPSC lines demonstrated regular microglia morphology with many sinus branches stemming from a small cell body (Figure 5B). Next, we performed RT-qPCR experiments to compare the transcriptomics profiles of microglia -like cells from hiPSC-line-1 and 2, generated manually, to those generated using the CompacT SelecT robot, in addition to hiPSCs and pre-macrophage precursors. Microglia-like cells derived from two different hiPSC lines using both manual and automated procedures exhibited similar gene expression patterns for multiple microglia markers like APOE, GRP34, TREM2, CD33, MERTK, and ITGAM (Figure 5C). IBA1 is a microglia-specific marker, which has been shown to be upregulated during phagocytosis.³⁸ To determine the functionality of our two hiPSCs-differentiated microglia lines, we stained them for IBA1 and evaluated their phagocytic ability to internalize pHrodo green-stained E. coli cells after LPS stimulation. As shown in Figure 5D, immunostaining confirmed that both microglia lines are positive for IBAI and internalized pHrodo green E. coli after LPS stimulation. The phagocytic ability of these microglia cells was inhibited when treated with Cytochalasin D (Figure 5E), conforming to the specificity of the phagocytic ability of these cells. Thus, hiPSCs-differentiated microglia cells using the weekend-free protocols are comparatively similar both transcriptionally and functionally in manual and automated differentiations.

Figure 5.

Characterization of microglia-like cells using weekend-free protocol manually and by robotics. (A) Overview of microglia differentiation protocol. (B) Phase-contrast image showing typical microglia-like cells on day 7. Scale bar, 120 μm. (C) Relative mRNA expression levels of APOE, GRP34, TREM2, CD33, MERTK, and ITGAM in hiPSC at D0 and differentiated hiPSC-derived cells. Values are normalized first to 18S and then to hiPSC at D0, with hiPSC at D0 serving as the baseline. Data are presented as mean ± SEM from 3 distinct biological samples, with individual values indicated by different symbols. Groups include hiPSC-line 1 (manual and automated protocols) and hiPSC-line 2 (manual and automated protocols). (D) Microglia-like cells at day 7, immunostained for IBA1 (green). Scale bar 120 and 230 µm. (E) Microglia-like cells phagocytosis. Microglia-like cells phagocytose pHrodo green upon LPS challenge. Cytochalasin D inhibits pHrodo green uptake in microglia-like cells upon LPS challenge.

Scalable production for retinal pigment epithelial cells

Derivation of retinal pigment epithelial (RPE) cells at large scale has been challenging. RPE differentiation and maturation is lengthy, requires daily media changes for at least the first 20 days, to overcome these challenges, we adapted for scaling a previously published protocol to differentiate hiPSCs into RPE cells as outlined in Figure 6A.³⁹ To validate the usage of our optimized hiPSCs maintenance and RPE differentiation protocols, we decided to include a hiPSC line generated from the AREDS2-NEI Cohort and obtained from a repository at the NYSCF. The line is hiPSC-3 (BB0045-01-MSV-086). Phase contrast images show the distinctive hexagonal-shaped epithelial morphology of the manual and robotic hiPSCs-differentiated RPE cells as described previously (Figure 6B).⁴⁰ A comparison of manually and robotically differentiated RPE cultures from two D24 differentiated hiPSC lines indicated similar gene expression levels of typical RPE markers, CRALBP, MITF, BEST1, and MERTK as represented in Figure 6C. The similarity of the manual and robotic RPE differentiation protocols was further corroborated by measuring the protein expression levels of BEST1, RPE65, and ZO1 marker. Our data demonstrated that majority of cells express RPE-specific markers irrespective of manual or robotic differentiation (Figure 6D).^41,42 As depicted in Figure 6E, we observed no significant differences in the TEER measurements recorded in hiPSC lines-1 and 3 in manual versus automated differentiations. Overall, data presented here show no significant differences between cells differentiated manually or on the robot.

Figure 6.

Characterization of retinal pigment epithelial-like cells using both manual and automated protocols. (A) Overview of the retinal pigment epithelial cell differentiation protocol. (B) Phase-contrast image showing typical retinal pigment epithelial cells (day 66). Scale bar, 200 μm. (C) Relative mRNA expression levels of CRALBP, MITF, BEST1, and MERTK in hiPSC at D0 and differentiated hiPSC-derived cells. Values are normalized first to 18S and then to hiPSC at D0, with hiPSC at D0 serving as the baseline. Data are presented as mean ± SEM from 3 distinct biological samples, with individual values indicated by different symbols. Groups include hiPSC-line 1 (manual and automated protocols) and hiPSC-line 3 (manual and automated protocols). (D) Retinal pigment epithelial cells, immunostained for BEST1, RPE65, and ZO1. Scale bar, 50 µm. (E) Measurement of trans-epithelial electrical resistance (TEER) to show barrier integrity, TEER values calculated in ohm-cm² during RPE differentiation. ns: nonsignificant. Abbreviations: RIM: retinal induction media, RDM: retinal differentiation media, RMNA: retinal media with nicotinamide and activin-A, RPE-MM: RPE maturation media, MACS: magnetic activated sorting.

Discussion

In this study, we have developed and outlined an efficient, scalable, high-throughput, easily adaptable, weekend-free hiPSC cell culture protocol that can be used for a wide range of hiPSCs applications, including hiPSCs maintenance, biobanking, and differentiation into multiple different cell lineages. Moreover, our protocols have been adapted to automated robotic systems and provide multiple freezing time points, allowing experimental flexibility and mitigating the risk of losing precious hiPSC lines in case of any potential mishaps. This modified approach significantly simplifies the scaling of hiPSCs needed in different applications, such as disease modeling and drug testing.

In the past decade, hiPSCs have emerged as one of the most valuable, translatable preclinical in vitro human disease models,^1,22,43 due to their self-renewing capacity, the ability to differentiate into numerous and diverse cell types, feasibility of gene editing, and most importantly, the capability of generating these lines form patient-specific somatic cells for potential use in population-based studies and disease modeling. This is further emphasized by the Food and Drug Administration (FDA) Modernization Act 2.0, authorizing the use of alternative methods to animal testing, including hiPSC assays and artificial intelligence, to obtain an exemption from the FDA to investigate drug safety and effectiveness. The scale of these studies will require the maintenance of hundreds or thousands of multiple different patient-derived hiPSC lines and the differentiation of these cells into different cell lineages for various 2D or 3D functional assays. In this study we have developed and optimized a scalable, weekend-free hiPSC maintenance and amplification protocol and benchmarked it to a conventional maintenance protocol by characterizing the hiPSCs cells for its morphology, pluripotency gene expression, and differentiation capabilities. All hiPSC lines maintained using our manual and automated weekend-free protocol maintained their pluripotency potential and were successfully differentiated into multiple extremely challenging cell lineages like microglia and RPEs. Moreover, we further verified the effectiveness of our protocols by amplifying a cohort of 72 hiPSC lines in less than 6 months for use in disease modeling and drug testing experiments. All our maintenance and differentiation protocols were smoothly transferred to CompacT SelecT, a robotics system enabling us to generate hiPSCs and/or differentiated cells of consistently high reproducible quality and significantly higher yields. Overall, we demonstrate the effectiveness and scalability of the weekend-free optimized protocol for hiPSCs maintenance and differentiation both manually and using automation.

Some limitations of our study may include restrictive access to robotics, especially in academic labs where funding may be a limited resource. However, in recent years, a variety of automated cell culture systems, each offering distinct features, throughput capacities, vessel compatibility, and flexible pricing models are available such as Hamilton’s STAR, Molecular Devices CellExpress.ai, Formulatrix’s CELLMATIC and Celltrio’s RoboCell. Some systems are designed for plate-based workflows and high throughput capacities, while others focus on larger-scale flask or bioreactor cultures. This provides more choices for labs with different budgets, lab infrastructure, and personnel expertise and should consider factors like scalability, operational flexibility, protocol customization, technical support, and integration with downstream assays to determine the most suitable option. Published reports have described successful use of these systems for hiPSC maintenance, amplification, and differentiation.³⁴ Our work complements existing literature by providing a robust weekend-free manual protocol that is automation-compatible but does not require specialized equipment and ensuring accessibility. For labs seeking automation, we recommend a careful evaluation of current robotic platforms based on throughput, technical requirements, support, and cost. As automated technologies continue to advance, we anticipate wider adoption and ongoing improvements in reproducibility, scalability, and efficiency for pluripotent stem cell culture. Moreover, we would like to acknowledge that while automation requires upfront investment, it offers long-term advantages by lowering labor costs, freeing valuable resources for applying the models, enhancing consistency and reproducibility, minimizing risks of error and contamination, and enabling true high-throughput processing. As shown by our amplification of 72 hiPSC lines in under 6 months, automation delivers major gains in speed and research output for large-scale studies. Importantly, our weekend-free manual protocols remain highly accessible for labs unable to adopt automation allowing high-quality, scalable hiPSC maintenance in any settings. Our work is meant to provide practical solutions for both automated and manual workflows, this allows any lab to choose an approach that fits their resources and research needs. Second, we relied on single-cell passage and the use of ROCKi in our protocols, which has been anecdotally associated with chromosomal abnormalities. To overcome this challenge, we frequently karyotyped our cultured cells to check for the presence of any chromosomal abnormalities.

Indeed, hiPSC protocols remain challenging, labor intensive, expensive, and require daily media changes, which hinders the capacity to standardize and scale hiPSC work and requires work over the weekends.^1,22,43,44 Several labs have tested and published weekend-free suitable hiPSCs maintenance protocols using split-ratio for cell seeding³⁵; however, that approach did not account for potential differences in pre-splitting densities and variabilities in growth rates between different lines.^25,30–33 Our optimized weekend-free protocol offers a unique ability to maintain and differentiate hiPSC lines in a predictable, defined, and high-throughput manner. Our maintenance protocol relies exclusively on defined cell seeding density, a factor once normalized will help synchronize the growth of all cell lines in culture. This protocol is highly flexible and is scalable over a wide range of culture vessel formats. We relied on single-cell passaging, Matrigel coating, and mTeSR Plus media. However, our lab has been using this protocol with multiple passaging reagents such as Versene/EDTA, different types of culturing media such as E8, Stemflex, and mTeSR, and different coating reagents including Matrigel, Geltrex, and Vitronectin.³⁵ Furthermore, to ensure consistency in our cell culture experiments, we implemented the following quality control step: prior to large-scale cell plating on the CompacT Select, we perform a test run using small batches of Matrigel and media 24–48 h in advance. Importantly, this test run is carried out on CompactT Select, allowing us to monitor cell attachment and growth under conditions identical to the actual experiment. This approach enables us to adjust, if necessary, before proceeding with the full batch. Another advantage of our protocol is simplicity. HiPSC culture work requires extensive hands-on training to be able to handle cells in the most delicate way. Our protocols are flexible and define robust guidelines that require minimal training, making it suitable for high-throughput labs that are specialized in cell production and generation. To further expand the application of our protocols, we differentiated our hiPSC lines to various types of cells with varying complexity, ranging from endothelial cells to microglia and RPE cells. Differentiation protocols were modified from established published protocols and were further improved by making them weekend-free and automation compatible.^37,39 Coupling our maintenance and amplification protocols with differentiation protocols that offer freezing capabilities will open new frontiers in the hiPSC field. This work will (1) reduce the labor needed to maintain hiPSC lines by skipping 3 days of media changes, (2) help design well-synchronized experiments for multiple lines much needed for disease modeling and drug testing, (3) unify the timeline of differentiation protocols needed for co-culture and 3D modeling, and (4) alleviate the scalability of the hiPSCs work.

In conclusion, the manual and automated weekend-free optimized protocols presented in this study overcome many current challenges associated with hiPSC cell culture work. Our protocols enable the synchronized growth of multiple hiPSC lines and simultaneous differentiations of hiPSCS into multiple disease-relevant cell types. By improving the scalability, reproducibility, and efficiency of hiPSCs culture and differentiation, our protocols facilitate the production of consistent, high quality, and reliable cell types for use in both research and therapeutic applications.

Author contributions

M.J. and S.P. designed the research. M.J., S.P, D.D., G.G., E.Y.C., A.G., C.H., J.T., L.B., E.K., A.F., S.B., D.V., J.D.P., and J.G. completed experiments. K.B. and E.W. advised on imaging experiments, V.E.T and C.H.L advised on microglia experiments. A.D.H. and P.R. supervised the project. M.J., S.P., A.D.H. and P.R. wrote the manuscript with input from all the other authors.

Supplementary material

Supplementary material is available at Stem Cells Translational Medicine online.

Funding

None declared.

Conflicts of interest

All authors are or were employees of AbbVie. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication. No honoraria or payments were made for authorship. Statement: During the preparation of this work, the author(s) used the AbbVie Intelligence Platform to check for grammatical and spelling errors. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Data availability

The data underlying this article are available in the article and in its Supplementary Material.

Ethics approval statement

Ethics approval was not required for this work because none of the hiPSC lines used are internal; all lines were obtained from external sources. For NYSCF lines, we have permission for use based on the Material Transfer Agreement (MTA) that was reviewed and approved by the NYSCF Ethics Committee. For EBiSC and Coriell lines, all available MTAs were provided. The responsibility for obtaining ethics approval and informed consent for these cell lines lies with the original owners. Copies of these MTAs are available to be included as Supplementary Material if required.